Pollution monitoring method and system for river water and storage medium

Abstract

The invention relates to the technical field of pollution monitoring, in particular to a pollution monitoring method and system for riverway water and a storage medium, and the method comprises the steps: obtaining the spectral data of riverway sediment through a multi-point synchronous collection technology, extracting a feature vector through principal component analysis, and determining the types of pollutants; in combination with hydrodynamic force and molecular polarization effects, a back propagation neural network is utilized to predict pollutant diffusion direction and interface polarization intensity, and a distribution mode of pollutants in sediments is generated; obtaining potential gradient and chemical bond amplitude characteristics of pollution source space distribution; matching the signal attenuation rate of the spectral signal with a chemical bonding strength model, and determining a pollutant emission frequency and an interface scattering effect; and the spatial distribution, emission frequency and potential gradient of the pollution source are traced in combination with the sustainability of the pollution source and the amplitude of the chemical bond. Through the steps, accurate tracing and dynamic monitoring of the pollution source are realized, and the accuracy of pollutant detection and the scientificity of water quality treatment are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of pollution monitoring technology, and in particular to a pollution monitoring method, system and storage medium for river water. Background Art

[0002] Pollutant analysis and source tracing in river sediments presents a complex technical challenge. Traditional methods struggle to accurately identify the types, sources, and dynamic patterns of multiple pollutants. The diffusion of pollutants in river channels is influenced by hydrodynamic conditions and involves molecular polarization effects, which complicates analysis. The spatial distribution of pollution sources exhibits characteristics of potential gradients and chemical bond amplitudes, but extracting effective information from these characteristics and correlating them with pollutant emission frequencies remains a challenge. Furthermore, the temporal dynamics of pollutants and the persistence of pollution sources require further study. Crucially, integrating these complex physicochemical processes with spectral data analysis methods to establish a systematic technical framework for pollutant analysis and source tracing in river sediments is crucial. This requires not only identifying individual pollutants but also considering the interactions among multiple pollutants and their migration and transformation patterns within river ecosystems. Furthermore, ensuring the accuracy and reliability of analytical results to provide strong support for water environment management decisions is a pressing issue. Summary of the Invention

[0003] The present invention provides a method for monitoring river water pollution, which mainly includes: Step S1: Acquire river sediment spectral data, extract a first eigenvector based on principal component analysis, and determine the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectral template; Step S2: determining the pollutant diffusion direction and interface polarization intensity through a back propagation neural network according to the pollutant type, hydrodynamic influence, and molecular polarization effect, and obtaining the distribution pattern of the pollutants in the sediment; Step S3: Based on the distribution pattern and wavelength scattering intensity, cluster analysis is performed to obtain the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the pollution source. The signal attenuation rate of the spectral signal is matched with the preset chemical bond strength model to determine the pollutant emission frequency and interface scattering effect. Step S4: obtaining a dynamic trend of the pollutant pulse duration and the pollutant deposition ratio in the sediment by analyzing the time series of the pollutant emission frequency and the interface scattering effect; Step S5: Determine the persistence of the pollution source based on the deviation between the dynamic trend and the preset pollutant diffusion model, and obtain the potential gradient distribution of the spatial distribution of the pollution source and the traceability results of the pollutant emission frequency through weighted averaging.

[0004] As a preferred technical solution of the present invention, in step S1, obtaining river sediment spectral data includes: Through multi-point synchronous acquisition, spectral data are acquired from surface and deep sediments of the river channel to obtain an original spectral dataset containing spectral phase shift, time delay distribution, and wavelength absorption coefficient. The original spectral dataset is preprocessed to extract the characteristic parameters of the spectral signal, where the characteristic parameters include spectral phase shift, time delay distribution, and wavelength absorption coefficient. The initial characteristic matrix of the spectral dataset is constructed based on the characteristic parameters, and the initial characteristic matrix is ​​denoised to obtain an optimized spectral dataset.

[0005] As a preferred technical solution of the present invention, extracting the first eigenvector by principal component analysis includes: The original spectral data set is standardized to obtain a standardized spectral matrix. The standardized spectral matrix is ​​decomposed by principal component analysis to extract the first eigenvector containing spectral phase shift, surface charge density, and time delay distribution. The variance contribution rate of the first eigenvector is calculated, and the principal components are screened according to the variance contribution rate to obtain the principal component feature set. The principal component feature set is orthogonalized to generate the optimized first eigenvector.

[0006] As a preferred technical solution of the present invention, in step S1, determining the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectrum template includes: Obtain a preset pollutant spectral template, calculate the cosine similarity between the first eigenvector and the preset pollutant spectral template, compare the cosine similarity with the preset threshold, and if the cosine similarity is greater than the preset threshold, determine the type of pollutant. Extract the spectral characteristic parameters of the determined pollutant type and generate a pollutant characteristic vector. Verify the accuracy of the pollutant type based on the matching result between the pollutant characteristic vector and the preset pollutant spectral template.

[0007] As a preferred technical solution of the present invention, in step S2, the pollutant diffusion direction and interface polarization strength are determined by a back propagation neural network, including: An input feature set is constructed based on the first eigenvector corresponding to each type of pollutant, and a training data set is generated by combining the hydrodynamic influence and molecular polarization effect. The training data set is trained through a back-propagation neural network to obtain a pollutant diffusion prediction model. The pollutant diffusion prediction model is used to calculate the pollutant diffusion direction and interface polarization intensity, and the distribution pattern of pollutants in the sediment is generated. The distribution pattern is verified to obtain the spatial characteristics of the pollutant distribution.

[0008] As a preferred technical solution of the present invention, in step S3, the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the spatial distribution of the pollution source are obtained through cluster analysis, including: Spatial characteristic parameters are extracted based on the distribution pattern of pollutants in sediments. A characteristic data set is generated by combining the time delay distribution and wavelength scattering intensity. The characteristic data set is classified through cluster analysis to obtain the potential gradient distribution of the spatial distribution of pollution sources. The spatial characteristics of the chemical bond amplitude are calculated, and the potential gradient distribution and chemical bond amplitude are standardized to generate a characteristic matrix of the spatial distribution of pollution sources.

[0009] As a preferred technical solution of the present invention, in step S3, matching the signal attenuation rate of the spectral signal with a preset chemical bonding strength model includes: Obtain the signal attenuation rate of the spectral signal, extract the preset chemical bond strength model, calculate the matching degree between the signal attenuation rate and the preset chemical bond strength model, compare the matching degree with the preset threshold, and if the matching degree is greater than the preset threshold, determine the pollutant emission frequency and the interface scattering effect, perform spectral analysis on the pollutant emission frequency, generate characteristic parameters of the interface scattering effect, and verify the accuracy of the pollutant emission frequency.

[0010] As a preferred technical solution of the present invention, the potential gradient distribution of the spatial distribution of pollution sources and the traceability results of pollutant emission frequencies are obtained by weighted averaging, including: Characteristic parameters are extracted based on the persistence of the pollution source and the chemical bond amplitude, and a weighted feature set is generated by combining the potential gradient distribution and wavelength modulation characteristics. The weighted feature set is processed by weighted averaging to obtain the potential gradient distribution of the spatial distribution of the pollution source. The tracing results of the pollutant emission frequency are calculated, and the tracing results are verified to generate the final characteristic matrix of the spatial distribution of the pollution source.

[0011] In a second aspect, the present invention further provides a pollution monitoring system for river water, for implementing the above method, the system comprising: a determination unit, configured to obtain river sediment spectral data, extract a first eigenvector based on principal component analysis, and determine the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectral template; A prediction unit is used to determine the diffusion direction of pollutants and the intensity of interface polarization based on the pollutant type, hydrodynamic influence and molecular polarization effect through a back propagation neural network to obtain the distribution pattern of pollutants in the sediment; A matching unit is used to obtain the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the spatial distribution of the pollution source through cluster analysis based on the distribution pattern and wavelength scattering intensity, and to match the signal attenuation rate of the spectral signal with the preset chemical bond strength model to determine the pollutant emission frequency and interface scattering effect; An analysis unit, configured to obtain a dynamic trend of a pollutant pulse duration and a pollutant deposition ratio in sediments by analyzing a time series of the pollutant emission frequency and the interface scattering effect; The tracing unit is used to judge the persistence of the pollution source based on the deviation between the dynamic trend and the preset pollutant diffusion model, and obtain the tracing results of the potential gradient distribution of the spatial distribution of the pollution source and the pollutant emission frequency through weighted averaging.

[0012] In a third aspect, the present invention further provides a computer-readable storage medium having instructions stored thereon, and the instructions implement the above method when executed by a processor.

[0013] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention acquires river sediment spectral data through multi-point synchronous acquisition technology, extracts characteristic vectors through principal component analysis, and determines the type of pollutant based on the similarity with the preset template. Combining the hydrodynamic and molecular polarization effects, a back-propagation neural network is used to determine the diffusion direction of pollutants and the intensity of interface polarization, and obtain the pollutant distribution pattern. The potential gradient and chemical bond amplitude characteristics of the spatial distribution of pollution sources are obtained through cluster analysis, and the frequency of pollutant emissions is determined by matching the spectral signal attenuation rate with the chemical bond strength model. The dynamic trend of pollutants is obtained through time series analysis, the persistence of pollution sources is judged, and finally the distribution of pollution sources and emission frequency are traced using a weighted average algorithm. The present invention realizes the precise analysis and tracing of pollutants in river sediments, providing effective technical support for water environment governance. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Flowchart of a method for monitoring river water pollution according to an embodiment of the present invention; Figure 2 This is a structural diagram of a pollution monitoring system for river water in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The following will describe the technical solutions in the embodiments of the present invention in detail with reference to the accompanying drawings. The described embodiments are only a part of the embodiments of the present invention.

[0016] like Figure 1 The pollution monitoring method for river water in this embodiment may specifically include: Step S1: Acquire river sediment spectral data, extract a first eigenvector based on principal component analysis, and determine the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectral template; Among them, obtaining river sediment spectral data includes: Through multi-point synchronous acquisition, spectral data are acquired from surface and deep sediments of the river channel to obtain an original spectral dataset containing spectral phase shift, time delay distribution, and wavelength absorption coefficient. The original spectral dataset is preprocessed to extract the characteristic parameters of the spectral signal, where the characteristic parameters include spectral phase shift, time delay distribution, and wavelength absorption coefficient. The initial characteristic matrix of the spectral dataset is constructed based on the characteristic parameters, and the initial characteristic matrix is ​​denoised to obtain an optimized spectral dataset.

[0017] Specifically, spectral data of surface and deep sediments of the river are obtained through multi-point synchronous acquisition, including characteristics such as spectral phase shift, time delay distribution and wavelength absorption coefficient. These original spectral data can reflect the distribution characteristics of pollutants in different sedimentary layers. The collected spectral data are preprocessed to extract characteristic parameters containing key information, such as spectral phase shift, time delay distribution and wavelength absorption coefficient. These parameters can reflect the physical and chemical properties of pollutants. Based on the above characteristic parameters, an initial characteristic matrix is ​​constructed to organize these spectral characteristic parameters into a matrix format suitable for further analysis in order to process the data. In order to remove the interference of noise in the acquisition process on the analysis results, the initial characteristic matrix is ​​preprocessed and denoised to obtain an optimized spectral data set. The above technical scheme can effectively extract the characteristic information of pollutants and eliminate the influence of environmental noise on the data, providing high-quality input data for subsequent pollutant type determination, diffusion model analysis, etc., and more accurately reflecting the true distribution of pollutants in sediments, thereby providing a reliable basis for pollution source tracing and water quality management decision-making.

[0018] Furthermore, the first eigenvector is extracted by principal component analysis, including: The original spectral data set is standardized to obtain a standardized spectral matrix. The standardized spectral matrix is ​​decomposed by principal component analysis to extract the first eigenvector containing spectral phase shift, surface charge density, and time delay distribution. The variance contribution rate of the first eigenvector is calculated, and the principal components are screened according to the variance contribution rate to obtain the principal component feature set. The principal component feature set is orthogonalized to generate the optimized first eigenvector.

[0019] Specifically, by standardizing the original spectral data set, the absorption coefficient data of different wavelengths are adjusted to a unified scale so that the differences between different data dimensions will not affect subsequent analysis; the standardized spectral matrix is ​​decomposed using the PCA algorithm. PCA projects the original data into a new feature space through linear transformation, extracts several principal components reflecting the main variations of the data, and screens out key features such as spectral phase shift, surface charge density and time delay distribution; the variance contribution rate of each principal component reflects its importance in the data. The principal components screened based on the variance contribution rate can effectively retain most of the effective information in the data and remove noise; the screened principal component feature set is orthogonalized to eliminate the correlation between the principal components and ensure their linear independence, thereby improving the stability and accuracy of subsequent analysis, obtaining the optimized first eigenvector and using it as the basic input for pollutant detection, which can help accurately identify and quantitatively analyze different types of pollutants. The above technical solution, through the combination of standardization and PCA, effectively suppresses the noise in the spectral data and extracts the key features of the pollutant distribution, thereby improving the accuracy and efficiency of monitoring through subsequent pollution source tracing and diffusion prediction.

[0020] Furthermore, determining the type of pollutant according to the cosine similarity between the first eigenvector and a preset pollutant spectrum template includes: Obtain a preset pollutant spectral template, calculate the cosine similarity between the first eigenvector and the preset pollutant spectral template, compare the cosine similarity with the preset threshold, and if the cosine similarity is greater than the preset threshold, determine the type of pollutant. Extract the spectral characteristic parameters of the determined pollutant type and generate a pollutant characteristic vector. Verify the accuracy of the pollutant type based on the matching result between the pollutant characteristic vector and the preset pollutant spectral template.

[0021] Specifically, by calculating the cosine similarity between the first eigenvector and these spectral templates, cosine similarity is a method to measure the degree of similarity between two vectors in direction. The closer the value is to 1, the higher the similarity between the two vectors. If the calculated cosine similarity is greater than a preset threshold, the pollutant type can be determined to be the pollutant represented by the template; after determining the type of pollutant, the spectral characteristic parameters related to the pollutant type are extracted to generate a pollutant characteristic vector, which integrates the absorption characteristics, phase changes and other spectral information of the pollutant at different wavelengths for subsequent pollutant analysis; by matching the pollutant characteristic vector with the preset pollutant spectral template again, the accuracy of the pollutant type is verified, and the reliability of the identification result is further ensured; the above technical solution, by combining the feature extraction of spectral data with the cosine similarity algorithm, can accurately identify the type of pollutants in complex river water pollution monitoring, solving the difficulties of traditional methods in identifying multiple pollutants and tracing pollution sources.

[0022] Step S2: determining the pollutant diffusion direction and interface polarization intensity through a back propagation neural network according to the pollutant type, hydrodynamic influence, and molecular polarization effect, and obtaining the distribution pattern of the pollutants in the sediment; Among them, the back propagation neural network is used to determine the diffusion direction of pollutants and the interface polarization intensity, including: An input feature set is constructed based on the first eigenvector corresponding to each type of pollutant, and a training data set is generated by combining the hydrodynamic influence and molecular polarization effect. The training data set is trained through a back-propagation neural network to obtain a pollutant diffusion prediction model. The pollutant diffusion prediction model is used to calculate the pollutant diffusion direction and interface polarization intensity, and the distribution pattern of pollutants in the sediment is generated. The distribution pattern is verified to obtain the spatial characteristics of the pollutant distribution.

[0023] Specifically, the core of step S2 is to use a back-propagation neural network (BP neural network) to determine the diffusion direction and interfacial polarization strength of pollutants, and then use this information to generate the distribution pattern of pollutants in the sediment. This process first involves a comprehensive analysis of pollutant types, hydrodynamic influences, and molecular polarization effects. The neural network model is then used to process and predict these factors, and then the spatial distribution characteristics of the pollutants are derived.

[0024] Specifically, the first eigenvector corresponding to each type of pollutant is extracted through the spectral data obtained from river sediments. This eigenvector is generated by the principal component analysis method and is the key data for inputting into the neural network. Based on the above first eigenvector and combined with the hydrodynamic influence and molecular polarization effect, these eigenvectors and hydrological environmental information, such as water flow velocity, turbulence intensity and charge distribution of pollutant molecules, are integrated into a complete input feature set; by simulating the diffusion behavior of different types of pollutants under a variety of hydrodynamic and molecular polarization effects, the labeling of the data set is established. During the training process, the neural network adjusts its network weights according to these input feature data, and gradually learns the relationship between the diffusion pattern of pollutants and the interface polarization intensity. Through back propagation, its prediction error is continuously optimized, so that the network can accurately predict the diffusion direction of pollutants under different hydrological conditions; after training, back propagation The back propagation neural network will generate a pollutant diffusion prediction model, which can use known input features to predict the diffusion direction and interface polarization intensity of pollutants in sediments. Among them, the diffusion direction and interface polarization intensity are two key factors describing how pollutants are distributed in river sediments. They reflect the migration path of pollutants and the interaction between pollutants and sediment interfaces. Through this prediction model, the distribution pattern of pollutants in sediments can be generated, which shows the distribution of pollutant concentrations at different spatial locations. In this process, the back propagation neural network can not only predict the migration route of pollutants based on the differences in hydrodynamics and molecular polarization effects, but also accurately analyze the deposition of pollutants in sediments. This distribution pattern can be further verified by comparing with field sampling data to verify the accuracy of the model prediction results and ensure that the spatial characteristics of pollutant distribution are consistent with the actual situation.

[0025] The above technical solution solves several core problems existing in traditional pollutant monitoring methods. First, traditional methods are often unable to simultaneously consider the combined influence of multiple factors such as pollutant type, river hydrodynamics, and molecular polarization effects. By introducing the back propagation neural network, these complex multivariate relationships are effectively modeled, which can more accurately predict the diffusion and deposition behavior of pollutants in the river. Secondly, this method can achieve real-time dynamic monitoring. By analyzing the diffusion pattern of pollutants, the direction and intensity changes of pollutant migration can be obtained in a timely manner, providing strong support for tracing the source of pollution. More importantly, the neural network-based model can be continuously optimized as more data is accumulated, improving prediction accuracy and having strong adaptability. For example, in actual applications, if the river is under strong water flow conditions, the diffusion direction of pollutants may be significantly affected by hydrodynamics, while under static water conditions, the molecular polarization effect may dominate the migration path of pollutants. The back propagation neural network can combine these two factors and predict the dynamic behavior of pollutants through the trained model, providing accurate data support for pollution source tracing and water environment management.

[0026] Step S3: Based on the distribution pattern and wavelength scattering intensity, cluster analysis is performed to obtain the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the pollution source. The signal attenuation rate of the spectral signal is matched with the preset chemical bond strength model to determine the pollutant emission frequency and interface scattering effect. Among them, the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the spatial distribution of pollution sources are obtained through cluster analysis, including: Spatial characteristic parameters are extracted based on the distribution pattern of pollutants in sediments. A characteristic data set is generated by combining the time delay distribution and wavelength scattering intensity. The characteristic data set is classified through cluster analysis to obtain the potential gradient distribution of the spatial distribution of pollution sources. The spatial characteristics of the chemical bond amplitude are calculated, and the potential gradient distribution and chemical bond amplitude are standardized to generate a characteristic matrix of the spatial distribution of pollution sources.

[0027] Specifically, spatial characteristic parameters are extracted based on the pollutant distribution pattern in river sediments. These characteristic parameters include factors such as pollutant concentration, distribution density, and the impact of water flow on pollutant diffusion. These data reflect the spatial distribution law of pollutants in sediments and can provide a basis for subsequent analysis. On this basis, the time delay distribution and wavelength scattering intensity are combined to generate a characteristic data set. The time delay distribution refers to the propagation delay of the light signal during the diffusion of pollutants, and the wavelength scattering intensity reflects the interaction intensity between the pollutant and the light signal. These data can provide information on the changes of pollutants in different sedimentary layers; after the data set is constructed, cluster analysis is used to classify the characteristic data to further reveal the spatial distribution of pollution sources. Cluster analysis can identify concentrated areas and locations of pollutants by grouping data according to similarity. The diffusion direction of pollutants is determined by the potential gradient distribution and the chemical bond amplitude, which form the spatial characteristics of the potential gradient distribution and the chemical bond amplitude. The potential gradient distribution can reflect the diffusion trend of pollutants in sediments, and the chemical bond amplitude provides information on the molecular level of pollutants, showing the chemical bonding strength between pollutants and sediment particles. These two characteristics jointly characterize the behavior and state of pollutants in river ecosystems. By standardizing the potential gradient distribution and the chemical bond amplitude, the two characteristics can be compared and analyzed at the same scale. Standardization helps to eliminate dimensional differences in the data and ensure the comparability between different characteristics, thereby generating a characteristic matrix of the spatial distribution of pollution sources. This matrix contains the spatial information of pollution sources, the diffusion intensity of pollutants and the intensity of their interaction with sediments, providing precise spatial positioning for tracing pollution sources and water body management.

[0028] The above-mentioned technical solution can realize efficient and accurate pollution source identification and distribution analysis during the pollutant monitoring process, and can reflect the pollutant diffusion trend and the persistence of the pollution source in real time, solving the problem that traditional monitoring methods are difficult to capture the changing laws of pollutants. By combining spectral data and physical and chemical characteristics, this method lays the foundation for pollutant source tracing, diffusion direction prediction and optimization of pollution control measures; for example, in the pollution monitoring of a certain river, by adopting cluster analysis technology, the specific location of the pollution source can be identified and the diffusion trend of the pollutant can be understood. When this technology is applied to water pollution control, managers can take targeted control measures according to the spatial distribution characteristics of the pollution source, such as adjusting water flow scheduling or setting up pollution source isolation zones, thereby optimizing the control effect.

[0029] Furthermore, matching the signal decay rate of the spectral signal with a preset chemical bonding strength model includes: Obtain the signal attenuation rate of the spectral signal, extract the preset chemical bond strength model, calculate the matching degree between the signal attenuation rate and the preset chemical bond strength model, compare the matching degree with the preset threshold, and if the matching degree is greater than the preset threshold, determine the pollutant emission frequency and the interface scattering effect, perform spectral analysis on the pollutant emission frequency, generate characteristic parameters of the interface scattering effect, and verify the accuracy of the pollutant emission frequency.

[0030] Specifically, spectral signals from river water are acquired through spectral acquisition technology. After a series of processing steps, these signals are used to calculate their signal attenuation rate. This rate reflects the degree of attenuation of the light signal in river water sediments and is generally closely related to the concentration, distribution, and properties of the pollutants. This attenuation rate is obtained by measuring the intensity changes of light waves as they propagate through the water column, combining the wavelength absorption coefficient and other optical parameters. A preset chemical bond strength model is also used. This model is pre-established based on the type of pollutant and sediment characteristics, and can describe the strength of the chemical bonds between pollutant molecules and sediments, and how this changes over time. This model reflects the emission characteristics of pollutants and their behavior in sediments, including processes such as pollutant diffusion, adsorption and interfacial reaction; this model provides a theoretical basis for analyzing the behavior of pollutants and can match the actual measured spectral signal attenuation rate; by calculating the matching degree between the attenuation rate of the spectral signal and the chemical bonding strength model, the distribution pattern of pollutants in river sediments can be evaluated. When the matching degree exceeds the preset threshold, the emission frequency of pollutants and the characteristics of the interfacial scattering effect can be determined. This process provides accurate data support for the monitoring and tracing of pollution sources, especially in the determination of the emission pattern and diffusion direction of pollutants, which can effectively improve the accuracy of detection results.

[0031] Among them, the pollutant emission frequency refers to the emission intensity of the pollution source within a specific period of time, which directly affects the concentration change of pollutants in the water body, and the interface scattering effect reflects the interaction between pollutants and the interface between water and sediment, which is usually related to the chemical properties, physical state and diffusion characteristics of the pollutants. By spectrally analyzing the emission frequency of pollutants, the time information about pollutant emissions can be extracted from the spectral data, providing a reference for the dynamic changes of pollution sources. This spectrum analysis can effectively reveal the short-term and long-term emission patterns of pollutants and support targeted pollution control. After determining the emission frequency, the accuracy of the frequency will be further verified. By combining actual river water environment data and historical pollution data, the verification step can eliminate errors and false positive results, ensure the reliability of pollution monitoring data, and analyze its correlation with environmental changes, flow changes and other factors by comparing pollutant emission data in different time periods, so as to accurately predict the dynamic change trend of pollutants.

[0032] The above technical solution solves the problem that the existing technology cannot monitor the emission frequency, diffusion direction and deposition distribution of pollutants in real time. By matching the spectral attenuation rate with the chemical bond strength model, it provides a more accurate means of pollutant monitoring. This solution can reflect the diffusion and change trend of pollutants in real time, which not only improves the accuracy of pollutant detection, but also makes the tracing and control of pollution sources more efficient.

[0033] Step S4: obtaining a dynamic trend of the pollutant pulse duration and the pollutant deposition ratio in the sediment by analyzing the time series of the pollutant emission frequency and the interface scattering effect; Specifically, since the frequency of pollutant emissions reflects the emission intensity of pollutants, and the interface scattering effect indicates the degree of scattering and adsorption of pollutants in sediments, the combination of the two can reveal the diffusion and accumulation process of pollutants in river water bodies. Therefore, by conducting time series analysis on these data, the duration of pollutant emissions can be determined, and then the changing trend of the deposition ratio of pollutants in river sediments can be inferred. This dynamic trend reflects the activity cycle, diffusion rate and deposition capacity of the pollution source, thereby providing a basis for tracing and controlling the pollution source.

[0034] By analyzing the time series of pollutant emission frequency and interface scattering effect, a dynamic model of pollutant diffusion is generated based on the data input of pollutant emission frequency and interface scattering effect, and the duration of pollutant emission is further derived. The above dynamic model is a model trained by the historical pollutant emission frequency, the corresponding interface scattering effect and the corresponding historical duration. The above analysis process is closely integrated with the diffusion behavior and deposition characteristics of pollutants, and can track the dynamic changes of pollutants in real time, so that the migration pattern, diffusion law and deposition rate of pollutants can be quantified, enhancing the timeliness and accuracy of pollution monitoring, and can effectively determine whether the pollution source is a persistent source and predict the potential risk and distribution pattern of pollutants, especially providing data support for water pollution monitoring and river pollution control. For example, in a certain river section, this technical solution can determine that when the emission frequency of a certain pollutant is high, its accumulation ratio in the sediment will increase. Conversely, the short-term change of the pollutant pulse indicates that the pollution source may be an intermittent emission source, thus providing data basis for environmental management departments to take corresponding control measures.

[0035] Step S5: Determine the persistence of the pollution source based on the deviation between the dynamic trend and the preset pollutant diffusion model, and obtain the potential gradient distribution of the spatial distribution of the pollution source and the traceability results of the pollutant emission frequency through weighted averaging.

[0036] Among them, the potential gradient distribution of the spatial distribution of pollution sources and the traceability results of pollutant emission frequency are obtained through weighted averaging, including: Characteristic parameters are extracted based on the persistence of the pollution source and the chemical bond amplitude, and a weighted feature set is generated by combining the potential gradient distribution and wavelength modulation characteristics. The weighted feature set is processed by weighted averaging to obtain the potential gradient distribution of the spatial distribution of the pollution source. The tracing results of the pollutant emission frequency are calculated, and the tracing results are verified to generate the final characteristic matrix of the spatial distribution of the pollution source.

[0037] Specifically, the above technical solution aims to use pollutant diffusion models and weighted averages to determine the persistence of pollution sources, and obtain the potential gradient distribution of the spatial distribution of pollution sources and the traceability results of pollutant emission frequencies through data processing. The implementation principle is based on the analysis of the dynamic trends and diffusion characteristics of river pollutants, combined with weighted average technology, and through the optimization and integration of different data features, to generate highly reliable spatial distribution information of pollution sources.

[0038] During the pollutant diffusion process, pollutants will diffuse with various factors such as water flow, sediment distribution and molecular polarization effect. Its diffusion pattern has temporal and spatial characteristics. The pollutant diffusion model is a model trained with training data. The training data includes historical pollutant spectral data, hydrodynamic effects, molecular polarization effects, interface polarization intensity, diffusion direction and corresponding pollutant distribution trends. When there is a large deviation between the detected dynamic trend and the predicted dynamic trend obtained by inputting the pollution diffusion model based on the current pollutant spectral data, hydrodynamic effects, molecular polarization effects, diffusion direction and interface polarization intensity, it can be inferred that the pollution source may have changed or its pollution emissions are more persistent. This deviation analysis will provide basic data for the weighted average algorithm to help further optimize the spatial distribution and emission characteristics of the pollution source; a weighted feature set is constructed by parameters such as chemical bond amplitude, potential gradient distribution and wavelength modulation characteristics. These feature parameters provide quantitative support for the distribution, migration and chemical reaction process of pollutants in sediments; for example, chemical bond amplitude can reflect the binding strength of pollutants and sediments, potential gradient distribution reveals the diffusion trend of pollutants under the action of water flow, and wavelength modulation characteristics further describe the pollution. The interaction between pollutants and spectral signals is integrated to improve the accuracy of describing the spatial distribution of pollution sources and accurately reflect the diffusion range and concentration gradient of pollutants in the river. The influence of different features is combined according to preset weights through weighted averaging to obtain the potential gradient distribution and pollutant emission frequency of the spatial distribution of pollution sources. This not only optimizes the processing effect of characteristic data, but also reduces the error caused by a single data source. The weighted averaging method can find a suitable balance between different features by rationally configuring weights. For example, the potential gradient has a greater weight when assessing the spatial extension of pollution sources, while the chemical bond amplitude can effectively distinguish the type of pollution sources. In this way, the weighted processing results will more accurately reflect the actual distribution and emission of pollution sources. By tracing the spatial distribution and emission frequency of pollution sources, a final characteristic matrix of pollution sources can be generated. This matrix contains the potential gradient and emission frequency of pollution sources at different time points and spatial locations, which can provide an accurate decision-making basis for pollutant control. By verifying the tracing results, the model can be further optimized, making the pollution source tracing results more reliable and providing scientific guidance for subsequent pollution source control and water quality improvement work.For example, in urban river pollution monitoring, the location of pollution sources can be accurately tracked and the frequency of pollutant emissions can be assessed. If a pollution source is located upstream of the river and frequently discharges heavy metal pollutants, the algorithm will combine potential gradient and chemical bond amplitude characteristics to generate a clear pollution source distribution map and determine its persistence through dynamic trends. In this case, the weighted average method can combine different characteristics such as water flow and the chemical properties of pollutants for precise analysis, ultimately forming a traceable result of the spatial distribution of pollution sources and emission frequency, thereby achieving precise monitoring and control of pollution sources.

[0039] The above technical solution solves the problems of inaccurate pollution source tracing and difficulty in judging the sustainability of pollution sources during pollutant monitoring by combining pollutant diffusion models, dynamic trend analysis and weighted averaging, and can achieve accurate positioning and dynamic monitoring of pollution sources.

[0040] The present invention also provides a pollution monitoring system for river water, which is used to implement the above method, such as Figure 2 As shown, the system includes: a determination unit, configured to obtain river sediment spectral data, extract a first eigenvector based on principal component analysis, and determine the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectral template; A prediction unit is used to determine the diffusion direction of pollutants and the intensity of interface polarization based on the pollutant type, hydrodynamic influence and molecular polarization effect through a back propagation neural network to obtain the distribution pattern of pollutants in the sediment; A matching unit is used to obtain the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the spatial distribution of the pollution source through cluster analysis based on the distribution pattern and wavelength scattering intensity, and to match the signal attenuation rate of the spectral signal with the preset chemical bond strength model to determine the pollutant emission frequency and interface scattering effect; An analysis unit, configured to obtain a dynamic trend of a pollutant pulse duration and a pollutant deposition ratio in sediments by analyzing a time series of the pollutant emission frequency and the interface scattering effect; The tracing unit is used to judge the persistence of the pollution source based on the deviation between the dynamic trend and the preset pollutant diffusion model, and obtain the tracing results of the potential gradient distribution of the spatial distribution of the pollution source and the pollutant emission frequency through weighted averaging.

[0041] The present invention also provides a computer-readable storage medium having instructions stored thereon, and the above-mentioned method is implemented when the instructions are executed by a processor.

[0042] In summary, the present invention uses multi-point synchronous acquisition technology to obtain spectral data of river sediments, combines principal component analysis to perform dimensionality reduction processing on the data, and extracts key features of pollutants. These features provide a high-quality data basis for subsequent identification and distribution analysis of pollutant types. By matching the cosine similarity with the preset pollutant spectral template, the type of pollutant can be accurately determined, providing a basis for further analysis of the distribution and diffusion of pollution sources; a back-propagation neural network is used in combination with hydrodynamic influences and molecular polarization effects to predict the diffusion direction of pollutants and the intensity of interface polarization, thereby obtaining the distribution pattern of pollutants in sediments. This step improves the accuracy of diffusion prediction by considering the interaction between water flow and pollutants and simulating the migration process of pollutants; in the analysis of the spatial distribution of pollution sources, a cluster analysis method is used in combination with the distribution pattern, potential gradient and chemical bond amplitude characteristics of pollutants to obtain a spatial distribution map of pollution sources. This step provides detailed spatial information for locating pollution sources, helping to identify high-risk areas; finally, the weighted average algorithm integrates various features, combines the persistence and dynamic trends of pollutants, accurately calculates the potential gradient distribution and emission frequency of pollution sources, and ultimately generates pollution source tracing results; the above technical solution can achieve real-time monitoring and evaluation of pollution sources, and judge changes in pollution sources through continuous dynamic trend analysis; through the coordination of various steps, a complete pollution source analysis chain is formed, from data collection to pollution source location, and then to dynamic monitoring of diffusion trends, achieving high-precision tracing and analysis of pollution sources. This technical solution solves the problem of difficult accurate identification and dynamic monitoring of pollution sources in traditional methods, and provides strong data support for water environment governance.

[0043] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A method for monitoring river water pollution, characterized in that: The method comprises: Step S1: Acquire river sediment spectral data, extract a first eigenvector based on principal component analysis, and determine the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectral template; Step S2: determining the pollutant diffusion direction and interface polarization intensity through a back propagation neural network according to the pollutant type, hydrodynamic influence, and molecular polarization effect, and obtaining the distribution pattern of the pollutants in the sediment; Step S3: Based on the distribution pattern and wavelength scattering intensity, cluster analysis is performed to obtain the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the pollution source. The signal attenuation rate of the spectral signal is matched with the preset chemical bond strength model to determine the pollutant emission frequency and interface scattering effect. Step S4: obtaining a dynamic trend of the pollutant pulse duration and the pollutant deposition ratio in the sediment by analyzing the time series of the pollutant emission frequency and the interface scattering effect; Step S5: Determine the persistence of the pollution source based on the deviation between the dynamic trend and the preset pollutant diffusion model, and obtain the potential gradient distribution of the spatial distribution of the pollution source and the traceability results of the pollutant emission frequency through weighted averaging.

2. The method according to claim 1, wherein Acquire river sediment spectral data, including: Through multi-point synchronous acquisition, spectral data are obtained from surface sediments and deep sediments in the river channel to obtain an original spectral dataset containing spectral phase shift, time delay distribution, and wavelength absorption coefficient. The original spectral dataset is preprocessed to extract the characteristic parameters of the spectral signal, where the characteristic parameters include spectral phase shift, time delay distribution, and wavelength absorption coefficient. The initial characteristic matrix of the spectral dataset is constructed based on the characteristic parameters. The initial characteristic matrix is ​​optimized using a noise reduction algorithm to obtain an optimized spectral dataset.

3. The method according to claim 1, wherein The first eigenvector is extracted through principal component analysis, including: The original spectral data set is standardized to obtain a standardized spectral matrix. The standardized spectral matrix is ​​decomposed through principal component analysis to extract the first eigenvector containing spectral phase shift, surface charge density, and time delay distribution. The variance contribution rate of the first eigenvector is calculated, and the principal components are screened according to the variance contribution rate to obtain the principal component feature set. The principal component feature set is orthogonalized to generate an optimized first eigenvector for subsequent pollutant type determination.

4. The method according to claim 1, wherein Determining the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectrum template includes: Obtain a preset pollutant spectral template, calculate the cosine similarity between the first eigenvector and the preset pollutant spectral template, compare the cosine similarity with the preset threshold, and if the cosine similarity is greater than the preset threshold, determine the type of pollutant. Extract the spectral characteristic parameters of the determined pollutant type and generate a pollutant characteristic vector. Verify the accuracy of the pollutant type based on the matching result between the pollutant characteristic vector and the preset pollutant spectral template.

5. The method according to claim 1, wherein The back propagation neural network is used to determine the diffusion direction of pollutants and the interface polarization intensity, including: An input feature set is constructed based on the first eigenvector, and a training data set is generated by combining the hydrodynamic influence and molecular polarization effect. The training data set is trained through a back-propagation neural network to obtain a pollutant diffusion prediction model. The pollutant diffusion prediction model is used to calculate the pollutant diffusion direction and interface polarization intensity, and the distribution pattern of pollutants in sediments is generated. The distribution pattern is verified to obtain the spatial characteristics of pollutant distribution.

6. The method according to claim 1, wherein The spatial characteristics of the potential gradient distribution and chemical bond amplitude of the pollution source spatial distribution are obtained through cluster analysis, including: Spatial characteristic parameters are extracted based on the distribution pattern of pollutants in sediments. A characteristic data set is generated by combining the time delay distribution and wavelength scattering intensity. The characteristic data set is classified through cluster analysis to obtain the potential gradient distribution of the spatial distribution of pollution sources. The spatial characteristics of the chemical bond amplitude are calculated, and the potential gradient distribution and chemical bond amplitude are standardized to generate a characteristic matrix of the spatial distribution of pollution sources.

7. The method according to claim 1, wherein Match the signal decay rate of the spectral signal to the preset chemical bond strength model, including: Obtain the signal attenuation rate of the spectral signal, extract the preset chemical bond strength model, calculate the matching degree between the signal attenuation rate and the preset chemical bond strength model, compare the matching degree with the preset threshold, and if the matching degree is greater than the preset threshold, determine the pollutant emission frequency and the interface scattering effect, perform spectral analysis on the pollutant emission frequency, generate characteristic parameters of the interface scattering effect, and verify the accuracy of the pollutant emission frequency.

8. The method according to claim 1, wherein The potential gradient distribution of the spatial distribution of pollution sources and the traceability results of pollutant emission frequencies are obtained through weighted averaging, including: Characteristic parameters are extracted based on the persistence of the pollution source and the chemical bond amplitude, and a weighted feature set is generated by combining the potential gradient distribution and wavelength modulation characteristics. The weighted feature set is processed by weighted averaging to obtain the potential gradient distribution of the spatial distribution of the pollution source. The tracing results of the pollutant emission frequency are calculated, and the tracing results are verified to generate the final characteristic matrix of the spatial distribution of the pollution source.

9. A pollution monitoring system for river water, used to implement the method according to any one of claims 1 to 8, characterized in that: The system comprises: a determination unit, configured to obtain river sediment spectral data, extract a first eigenvector based on principal component analysis, and determine the type of pollutant based on the cosine similarity between the first eigenvector and a preset pollutant spectral template; A prediction unit is used to determine the diffusion direction of pollutants and the intensity of interface polarization based on the pollutant type, hydrodynamic influence and molecular polarization effect through a back propagation neural network to obtain the distribution pattern of pollutants in the sediment; A matching unit is used to obtain the spatial characteristics of the potential gradient distribution and chemical bond amplitude of the spatial distribution of the pollution source through cluster analysis based on the distribution pattern and wavelength scattering intensity, and to match the signal attenuation rate of the spectral signal with the preset chemical bond strength model to determine the pollutant emission frequency and interface scattering effect; An analysis unit, configured to obtain a dynamic trend of a pollutant pulse duration and a pollutant deposition ratio in sediments by analyzing a time series of the pollutant emission frequency and the interface scattering effect; The tracing unit is used to judge the persistence of the pollution source based on the deviation between the dynamic trend and the preset pollutant diffusion model, and obtain the tracing results of the potential gradient distribution of the spatial distribution of the pollution source and the pollutant emission frequency through weighted averaging.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images